[br] Add audio extraction and support more URLs (Closes #2728)

This commit is contained in:
Sergey M․ 2014-04-09 20:19:27 +07:00
parent 29e3e682af
commit ef2041eb4e
1 changed files with 101 additions and 61 deletions

View File

@ -4,39 +4,72 @@ from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ExtractorError from ..utils import (
ExtractorError,
int_or_none,
)
class BRIE(InfoExtractor): class BRIE(InfoExtractor):
IE_DESC = "Bayerischer Rundfunk Mediathek" IE_DESC = 'Bayerischer Rundfunk Mediathek'
_VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?:[a-z0-9\-/]+/)?(?P<id>[a-z0-9\-]+)\.html$" _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-]+/)+(?P<id>[a-z0-9\-]+)\.html'
_BASE_URL = "http://www.br.de" _BASE_URL = 'http://www.br.de'
_TESTS = [ _TESTS = [
{ {
"url": "http://www.br.de/mediathek/video/anselm-gruen-114.html", 'url': 'http://www.br.de/mediathek/video/anselm-gruen-114.html',
"md5": "c4f83cf0f023ba5875aba0bf46860df2", 'md5': 'c4f83cf0f023ba5875aba0bf46860df2',
"info_dict": { 'info_dict': {
"id": "2c8d81c5-6fb7-4a74-88d4-e768e5856532", 'id': '2c8d81c5-6fb7-4a74-88d4-e768e5856532',
"ext": "mp4", 'ext': 'mp4',
"title": "Feiern und Verzichten", 'title': 'Feiern und Verzichten',
"description": "Anselm Grün: Feiern und Verzichten", 'description': 'Anselm Grün: Feiern und Verzichten',
"uploader": "BR/Birgit Baier", 'uploader': 'BR/Birgit Baier',
"upload_date": "20140301" 'upload_date': '20140301',
} }
}, },
{ {
"url": "http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html", 'url': 'http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html',
"md5": "ab451b09d861dbed7d7cc9ab0be19ebe", 'md5': 'ab451b09d861dbed7d7cc9ab0be19ebe',
"info_dict": { 'info_dict': {
"id": "2c060e69-3a27-4e13-b0f0-668fac17d812", 'id': '2c060e69-3a27-4e13-b0f0-668fac17d812',
"ext": "mp4", 'ext': 'mp4',
"title": "Über den Pass", 'title': 'Über den Pass',
"description": "Die Eroberung der Alpen: Über den Pass", 'description': 'Die Eroberung der Alpen: Über den Pass',
"uploader": None,
"upload_date": None
} }
} },
{
'url': 'http://www.br.de/nachrichten/schaeuble-haushaltsentwurf-bundestag-100.html',
'md5': '3db0df1a9a9cd9fa0c70e6ea8aa8e820',
'info_dict': {
'id': 'c6aae3de-2cf9-43f2-957f-f17fef9afaab',
'ext': 'aac',
'title': '"Keine neuen Schulden im nächsten Jahr"',
'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
}
},
{
'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
'info_dict': {
'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
'ext': 'mp4',
'title': 'Umweltbewusster Häuslebauer',
'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
}
},
{
'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
'md5': '23bca295f1650d698f94fc570977dae3',
'info_dict': {
'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
'ext': 'mp4',
'title': 'Folge 1 - Metaphysik',
'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
'uploader': 'Eva Maria Steimle',
'upload_date': '20140117',
}
},
] ]
def _real_extract(self, url): def _real_extract(self, url):
@ -44,56 +77,63 @@ class BRIE(InfoExtractor):
display_id = mobj.group('id') display_id = mobj.group('id')
page = self._download_webpage(url, display_id) page = self._download_webpage(url, display_id)
xml_url = self._search_regex( xml_url = self._search_regex(
r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL") r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
xml = self._download_xml(self._BASE_URL + xml_url, None) xml = self._download_xml(self._BASE_URL + xml_url, None)
videos = [] medias = []
for xml_video in xml.findall("video"):
video = {
"id": xml_video.get("externalId"),
"title": xml_video.find("title").text,
"formats": self._extract_formats(xml_video.find("assets")),
"thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")),
"description": " ".join(xml_video.find("shareTitle").text.splitlines()),
"webpage_url": xml_video.find("permalink").text
}
if xml_video.find("author").text:
video["uploader"] = xml_video.find("author").text
if xml_video.find("broadcastDate").text:
video["upload_date"] = "".join(reversed(xml_video.find("broadcastDate").text.split(".")))
videos.append(video)
if len(videos) > 1: for xml_media in xml.findall('video') + xml.findall('audio'):
media = {
'id': xml_media.get('externalId'),
'title': xml_media.find('title').text,
'formats': self._extract_formats(xml_media.find('assets')),
'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
'webpage_url': xml_media.find('permalink').text
}
if xml_media.find('author').text:
media['uploader'] = xml_media.find('author').text
if xml_media.find('broadcastDate').text:
media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
medias.append(media)
if len(medias) > 1:
self._downloader.report_warning( self._downloader.report_warning(
'found multiple videos; please ' 'found multiple medias; please '
'report this with the video URL to http://yt-dl.org/bug') 'report this with the video URL to http://yt-dl.org/bug')
if not videos: if not medias:
raise ExtractorError('No video entries found') raise ExtractorError('No media entries found')
return videos[0] return medias[0]
def _extract_formats(self, assets): def _extract_formats(self, assets):
def text_or_none(asset, tag):
elem = asset.find(tag)
return None if elem is None else elem.text
formats = [{ formats = [{
"url": asset.find("downloadUrl").text, 'url': text_or_none(asset, 'downloadUrl'),
"ext": asset.find("mediaType").text, 'ext': text_or_none(asset, 'mediaType'),
"format_id": asset.get("type"), 'format_id': asset.get('type'),
"width": int(asset.find("frameWidth").text), 'width': int_or_none(text_or_none(asset, 'frameWidth')),
"height": int(asset.find("frameHeight").text), 'height': int_or_none(text_or_none(asset, 'frameHeight')),
"tbr": int(asset.find("bitrateVideo").text), 'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
"abr": int(asset.find("bitrateAudio").text), 'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
"vcodec": asset.find("codecVideo").text, 'vcodec': text_or_none(asset, 'codecVideo'),
"container": asset.find("mediaType").text, 'acodec': text_or_none(asset, 'codecAudio'),
"filesize": int(asset.find("size").text), 'container': text_or_none(asset, 'mediaType'),
} for asset in assets.findall("asset") 'filesize': int_or_none(text_or_none(asset, 'size')),
if asset.find("downloadUrl") is not None] } for asset in assets.findall('asset')
if asset.find('downloadUrl') is not None]
self._sort_formats(formats) self._sort_formats(formats)
return formats return formats
def _extract_thumbnails(self, variants): def _extract_thumbnails(self, variants):
thumbnails = [{ thumbnails = [{
"url": self._BASE_URL + variant.find("url").text, 'url': self._BASE_URL + variant.find('url').text,
"width": int(variant.find("width").text), 'width': int_or_none(variant.find('width').text),
"height": int(variant.find("height").text), 'height': int_or_none(variant.find('height').text),
} for variant in variants.findall("variant")] } for variant in variants.findall('variant')]
thumbnails.sort(key=lambda x: x["width"] * x["height"], reverse=True) thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
return thumbnails return thumbnails